Data-efficient Neuroevolution with Kernel-Based Surrogate Models
This work addresses data-efficiency in neuroevolution for robotics and control tasks, representing an incremental improvement by adapting existing surrogate techniques to a specific bottleneck.
The paper tackled the challenge of applying surrogate models to neuroevolution, which is hindered by variable network topologies, by introducing a kernel-based surrogate model that uses NEAT's compatibility distance as a distance measure. The result showed that this algorithm achieved similar or better performance on cart-pole and half-cheetah tasks with several times fewer function evaluations compared to original NEAT.
Surrogate-assistance approaches have long been used in computationally expensive domains to improve the data-efficiency of optimization algorithms. Neuroevolution, however, has so far resisted the application of these techniques because it requires the surrogate model to make fitness predictions based on variable topologies, instead of a vector of parameters. Our main insight is that we can sidestep this problem by using kernel-based surrogate models, which require only the definition of a distance measure between individuals. Our second insight is that the well-established Neuroevolution of Augmenting Topologies (NEAT) algorithm provides a computationally efficient distance measure between dissimilar networks in the form of "compatibility distance", initially designed to maintain topological diversity. Combining these two ideas, we introduce a surrogate-assisted neuroevolution algorithm that combines NEAT and a surrogate model built using a compatibility distance kernel. We demonstrate the data-efficiency of this new algorithm on the low dimensional cart-pole swing-up problem, as well as the higher dimensional half-cheetah running task. In both tasks the surrogate-assisted variant achieves the same or better results with several times fewer function evaluations as the original NEAT.